Liveness.com
Biometric Liveness Detection Explained

 
 

What is “Liveness”?
 
In biometrics, Liveness Detection is an AI computer system’s ability to determine that it is interfacing with a physically present human being and not an inanimate spoof artifact.  
Note: It’s not called “Liveliness”. Don’t make that rookie mistake!
 


 
The History of Liveness
 

In 1950, Alan Turing (
wiki) developed the famous "Turing Test".  It measures a computer's ability to exhibit human-like behavior.  Conversely, Liveness Detection is AI that determines if a computer is interacting with a live human. 

Alan Turing
Turing c. 1928

 

The "Godmother of Liveness"
 

Dorothy E. Denning (
wiki) is a member of the
National Cyber Security Hall of Fame and coined the term “Liveness” in her 2001 Information Security Magazine Article: It's "liveness," not secrecy, that counts.  She states:

A good biometrics system should not depend on secrecy," and,

“... biometric prints need not be kept secret, but the validation process must check for liveness of the readings."

Decades ahead of her time, Dorothy E. Denning’s vision for Liveness Detection in biometric authentication could not have been more correct.

Dorothy E. Denning


Early Academic Papers About Liveness & Anti-Spoofing
 

One of the earliest papers on Liveness was published by Stephanie Shuckers, S.A., in 2002.  "Spoofing and anti-spoofing measures", and it is widely regarded as the foundation of today's academic body of work on the subject.  The paper states that "Liveness detection is based on recognition of physiological information as signs of life from liveness information inherent to the biometric".  

Later in 2016, her follow-up, "Presentations and Attacks, and Spoofs, Oh My", continued to influence presentation attack detection research and testing. 


 
Is Facial Recognition the Same as Liveness & Face Authentication?

No, and we all need to start using the correct terminology if we ever want to stop confusing people about biometrics! 

Facial Recognition is for surveillance; it's the 1-to-N matching of images captured with cameras the user doesn't control, like those in a casino or an airport. And it only provides "possible" matches for the surveilled person from face photos stored in an existing database. 

Face Authentication (1:1 Matching+Liveness), on the other hand, takes User-initiated data collected from a device they do control and confirms that User's identity for their own direct benefit, like, for example, secure account access.

They may share a resemblance and even overlap in some ways, but don't lump the two together. Like any powerful tech, this is a double-edged sword; Facial Recognition is a threat to privacy while Face Authentication is a huge win for it.

 

Should We Fear Centralized Face Authentication?
 

Fear of biometric authentication stems from the belief that centralized storage of biometric data creates a "honeypot" that, if breached, compromises the security of all other accounts that rely on that same biometric data.

Biometric detractors argue, "You can reset your password if stolen, but you can't reset your face."  While this is true, it is a failure of imagination to stop there.  We must ask, "What would make centralized biometric authentication safe?"

The answer is Certified Liveness Detection.  With it, the biometric honeypot is no longer something to fear because our security doesn't rely on our biometric data being kept secret.

Learn more about how Certified Liveness Makes Centralized Safe in this comprehensive FindBiometrics white paper.
 
 

How Liveness Detection Protects Us


Ms. Denning's photo posted above is biometric data, and is now cached on your computer.  Is she somehow more vulnerable now that you have it?  Not if her accounts are secured with Certified Liveness Detection, because that photo won't fool the AI.  Nor will a video, a copy of her driver license, passport, fingerprint, or iris.  She must be physically present to access her accounts, so she need not worry about keeping her biometric data "secret".

Liveness Detection prevents bots and bad actors from using stolen photos, deepfake videos, masks, or other spoofs to create or access online accounts.  Liveness ensures only real humans can create and access accounts.

Liveness checks solve some very serious problems.  For example, Facebook had to delete 5.4 billion fake accounts in 2019 alone!  Requiring proof of Liveness would have prevented these fakes from ever being created.
 

Note: In 2019, the crypto-currency wallet ZenGo offered a challenge: spoof Certified Liveness Detection and "steal" one Bitcoin (worth over $11,000 at the time).  A hi-res photo of the ZenGo CEO was provided, and the savvy cypherpunks gave it their best shot.  The ZenGo wallet remained unspoofed, and the bitcoin stayed safe, proving the efficacy of Certified Liveness Detection in one of the most public displays of biometric security to date (https://zengo.com/update-a-successful-zengo-challenge-for-us).

 

$75,000 Spoof Bounty Program

Don't be a guinea pig; insist that your Biometric Vendor maintain a persistent Spoof Bounty Program to ensure that they are aware of and robust to any emerging threats, like Deepfakes.  As of today, the only Biometric Authentication Vendor with an active, real-world Spoof Bounty is FaceTec.  Having now rebuffed over 8,000 Real-World Spoof Attacks, the goal of the $75,000 Spoof Bounty Program remains to uncover unknown vulnerabilities in the Liveness AI and Security scheme so they can be patched, and the anti-spoofing capabilities elevated even further.  Visit www.SpoofBounty.com to participate. 
 
 
Liveness for Onboarding, KYC and Enrollment

Requiring every new user to prove their Liveness before they are even asked to present an ID Document during digital onboarding is itself a huge deterrent to fraudsters who never want their real face on camera.
 
If an onboarding system has a weakness, the bad guys will exploit it to create as many fake accounts as possible.  To prevent this, Certified Liveness Detection during new account onboarding should be required.  Then we know that the new account belongs to a real human and their biometric data can be stored as a trusted reference of their digital identity in the future.

  

 
  
Liveness for Ongoing Authentication (Password Replacement)

 
Since most biometric attacks are spoof attempts, Certified Liveness Detection during user authentication must be mandatory.  With multiple high-quality photos of almost everyone available on Google or Facebook, a biometric authenticator cannot rely on secrecy for security. 

Liveness Detection is the first and most important line of defense against targeted spoof attacks on authentication systems.  The second is a very high FAR (see Glossary, below) for accurate biometric matching.   

With Certified Liveness Detection you can't even make a copy of your biometric data that would fool the system even if you wanted to.  Liveness catches the copies by detecting generation loss, and only the genuine physical user can gain access.




No Liveness Data = No Honeypot Risk
 
Two types of data are required for every Face Authentication: Face Data (for matching) and Liveness Data (to prove the Face Data was collected from a live person). 

Liveness Data must be timestamped, be valid only for a few minutes, and then deleted. Only Face Data should ever be stored.  New Liveness Data must be collected for every authentication attempt.  

Face photos are just "Face Data" without the corresponding Liveness Data, so they cannot be used to spoof Certified Liveness Detection, and thus, storing photos does not create honeypot risk.

Note: Think of the stored Face Data as the lock, the User's newly collected Face Data as a One-Time-Use key, and the Liveness Data as proof that key has never been used before. 


  
ISO/IEC 30107 - Liveness Testing Global Standard 


https://www.iso.org/standard/67381.html is the International Organization for Standardization’s (ISO) testing guidance for evaluation of Anti-Spoofing technology, a.k.a., Presentation Attack Detection (PAD).  Three document editions have been published to date, with a fourth edition currently in progress (as of November, 2019).
 
“bio” “metrics” literally means to measure live human physical characteristics.  Ironically, it took until late 2017 for anyone to release official guidance on how to determine if the subject of a biometric scan is actually alive.

Due to "hill-climbing" attacks (see Glossary, below), biometric systems should never reveal which part of the system did or didn't catch a spoof.  And while ISO 30107-3 gets a lot right, it unfortunately encourages testing both Liveness and Matching at the same time.  Scientific method requires the fewest variables possible be tested at once, so Liveness testing should be done with a solely Boolean (true/false) response.  Tests should not allow systems to have multiple-decision layers that could allow an artifact to pass Liveness but fail Matching because it didn't "look" enough like the enrolled subject. 
 
 
 
Spoof Artifact Levels
 

When a non-living object that exhibits human traits (an "artifact") is presented to a camera or biometric sensor, it's called a "spoof".  Photos, videos, masks, and dolls are all common examples of spoof artifacts.

  

 Artifact Level Description Example
 Level 1 (A)
 (iBeta testing available)
Hi-res paper & digital photos, digital deepfakes, hi-def challenge/response videos and paper masks.
 Level 2 (B)
 (iBeta testing available)
Commercially available lifelike dolls, and human-worn resin, latex & silicone 3D masks under $300 in price.
 
 Level 3 (C)
 (iBeta Does NOT test)
Custom-made ultra-realistic 3D masks, wax heads, etc., up to $3,000 in creation cost.

 
 

Non-Certified Liveness

Unfortunately, some types of Liveness Detection are uncertifiable because they are not secure enough to pass the lowest level of the ISO 30107 Presentation Attack Detection guidance requirements. 

Uncertifiable Liveness Detection methods include: blink, smile, turn/nod, colored flashing lights, making random faces, speaking random numbers, and many more. All easily spoofed.

User security and hard-won corporate credibility is put at risk by trusting unscrupulous vendor's exaggerated claims. 

When vendors claim to have "Robust Liveness Detection", they should "Pass the test or give it a rest!"

Note: Watch USAA Bank's non-certified "Facial Recognition" app security
get spoofed by a crude photo slideshow, easily unlocking
one of their user's bank accounts ------------------->

                
 

The Threat of Deepfakes
 
So-called "deepfakes" have been around for years, but now even the general public understands that digital media can be manipulated easily.

If non-certified Liveness Detection is vulnerable to deepfake spoofs derived from photos or videos, it cannot be used for biometric security. 

Note: Watch as a basic "deepfake" puppet is created in 20 seconds
that can be used to spoof almost every non-certified liveness
vendor on the market today ------------------->

                             

  

 
 

 

Certified Liveness & Compliant Liveness
 
NIST/NVLAP-accredited Lab, iBeta in Denver, CO USA, is the only Liveness testing lab guided by the ISO 30107 global testing standard (https://www.ibeta.com/biometric-testing).  Unfortunately, iBeta no longer provides comprehensive Certification testing, only time-limited compliance reviews.  iBeta has never provided Level 3 Certification testing, so FaceTec created a $75,000 Spoof Bounty Program to prove real-world Level 3 security.

All organizations have a fiduciary duty to provide the strongest Liveness Detection available to their users whenever remote biometric onboarding or authentication is required.

        
    
 Certified Face Liveness Vendors Non-Certified Vendors

Certified iBeta/NIST PAD: Level 1 & 2

       

Authenteq
BTS Digital
Certisign
e4 Global
EvidentID
FaceTec
Fractal
FintechOS
Fractal
Gemalto/Thales
Gulf Data-gDi
Idenfy
Jumio
Karalundi
Kvalifika
Neuvote
Ondato
OneyTrust
Passbase
PBSA Group
Polygon
Solus Connect
Sum & Substance
Synaps
TiC Now
TeraSystem
Valid
VerifyMyAge
Yoti
ZealiD

Iceland
Kazakhstan
Brazil
South Africa
USA
Bulgaria
USA
Romania
Germany
France
UAE
Lithuania
USA
Mexico
Georgia
Canada
Lithuania
France
USA
South Africa
Portugal
Singapore
United Kingdom
France
Chile
Philippines
Brazil
United Kingdom
United Kingdom
Sweden

   

$75,000 Spoof Bounty Program 




Aware
BioID
Daon
FacePhi
HID

IDWall
iProov
Imageware
Mati
Nuance

Sensory
& more...




Though they remain unnamed,
many vendors have tried and

failed iBeta's PAD testing.

   

Compliant Face Liveness Providers Non-Compliant Vendors
   
All 30107-3 Compliant Vendors: 
 
Idemia Level 1 & 2 France
ID R&D Level 1 USA
Incode Level 1 Mexico
OCR Labs
Level 1 Australia



No Spoof Bounty Programs


Most Other Liveness
Detection Vendors


Though they remain unnamed,
many vendors have tried and

failed iBeta's conformance program.

      

Compliant Fingerprint Liveness Providers Non-Compliant Vendors
  
    

HID Global

USA

All Vendors Compliant iBeta PAD: Level 1


 
No Spoof Bounty Program

 
All Other Fingerprint
Hardware Vendors


Though they remain unnamed,
many vendors have tried and

failed iBeta's conformance program.

    

 Compliant Palmprint Liveness Providers Non-Compliant Vendors
  
    

RedRock Biometrics

USA

All Vendors Compliant iBeta PAD: Level 1


 
No Spoof Bounty Program


Fujitsu
iProov


Though they remain unnamed,
many vendors have tried and

failed iBeta's conformance program.

 
   
  
Editors' Note: Should Liveness Detection Be Required By Law?
 

We believe that legislation must be passed to make Certified Liveness Detection mandatory if biometrics are used for Identity & Access Management (IAM).  Our personal data has already been breached, so we can no longer trust Knowledge Based Authentication (KBA).  We must turn our focus from maintaining databases full of "secrets" to securing attack surfaces.  Current laws already require organic foods to be certified, and every medical drug must be tested and approved.  In turn, governments around the world should require Certified Liveness Detection be used to protect the digital safety and biometric security of their citizens.


 
Editors' Note: Why 2D Face Matching doesn't work well on large data sets (500,000+)

We've all heard an actor say, "get my good side", and the best photographers know which distances and lenses make portrait photos the most flattering.  This is because a real 3D human face contains orders of magnitude more data than a typical 2D photo, and when a 3D face is flattened into a single 2D layer depth data is lost and creates significant issues.  In the real world, capture distance, camera position, and lens diameter play big parts in how well a derivative 2D photo represents the original 3D face.



Source – Best Portrait Lens – Focal Length, Perspective & Distortion Matt Granger – Oct 27, 2017

2D Face Matching will not always “see” her as the same person.  In some frames she might look more like her sister or her cousin and could match one of them even more highly than herself.  In large datasets these visual differences are within the margin of error of the 2D algorithms and they make confidence in the 1:N match impossible.  However, 3D FaceMaps not only provide more human signal for Liveness Detection, but they also provide data on the size and depth of the face which is combined with visual traits to increase accuracy, and enable the use of 1:N matching with significantly larger datasets.​


 

Editors' Note: The Problem With CAPTCHAs
 

CAPTCHA (wiki), an acronym for "Completely Automated Public Turing test to tell Computers and Humans Apart", is a simple challenge–response test used in computing to determine whether the user is human or a bot.

In an article on TheVerge.com, Josh Dzieza writes, “Google pitted one of its machine learning algorithms against humans in solving the most distorted text CAPTCHAs: the computer got the test right 99.8-percent of the time, while the humans got a mere 33 percent.” 

Jason Polakis, a computer scientist, used off-the-shelf image recognition tools, including Google's own image search, to solve Google's image CAPTCHA with 70% accuracy, states “You need something that’s easy for an average human, it shouldn’t be bound to a specific subgroup of people, and it should be hard for computers at the same time.”

Even without AI, services like: deathbycaptcha.com and anti-captcha.com allow bots to bypass the challenge–responses tests by using proxy humans to complete them.  With so many people willing to do this work, it's cheap to defeat at scale and workers earn between $0.25-$0.60 for every 1000 CAPTCHAs solved. (webemployed).


 

Resources & Whitepapers
 

Information Security Magazine - Dorothy E. Denning's (wiki) 2001 article, “It Is "Liveness," Not Secrecy, That Counts
 
FaceTec: There's a New Sheriff in Town - Standardized PAD Testing & Liveness Detection - Biometrics Final Frontier

Gartner, “Presentation attack detection (PAD, a.k.a., “liveness testing”) is a key selection criterion.  ISO/IEC 30107 “Information Technology — Biometric Presentation Attack Detection” was published in 2017.  
(Gartner’s Market Guide for User Authentication, Analysts: Ant Allan, David Mahdi, Published: 26 November 2018). FaceTec’s ZoOm was cited in the report.  For subscriber access: https://www.gartner.com/doc/3894073?ref=mrktg-srch.
 
Forrester, "The State Of Facial Recognition For Authentication - Expedites Critical Identity Processes For Consumers And Employees"  By Andras Cser, Alexander Spiliotes, Merritt Maxim, with Stephanie Balaouras, Madeline Cyr, Peggy Dostie.  For subscriber access: https://www.forrester.com/report/The+State+Of+Facial+Recognition+For+Authentication+And+Verification/-/E-RES141491#

Ghiani, L., Yambay, D.A., Mura, V., Marcialis, G.L., Roli, F. and Schuckers, S.A., 2017. Review of the Fingerprint Liveness Detection (LivDet) competition series: 2009 to 2015. Image and Vision Computing58, pp.110-128:
https://www.clarkson.edu/sites/default/files/2017-11/Fingerprint%20Liveness%20Detection%2009-15.pdf 

Schuckers, S., 2016. Presentations and attacks, and spoofs, oh my. Image and Vision Computing55, pp.26-30:
https://www.clarkson.edu/sites/default/files/2017-11/Presentations%20and%20Attacks.pdf

Schuckers, S.A., 2002. Spoofing and anti-spoofing measures. Information Security technical report(4), pp.56-62:
https://www.clarkson.edu/sites/default/files/2017-11/Spoofing%20and%20Anti-Spoofing%20Measures.pdf

  

 

Glossary - Biometrics Industry & Testing Terms:

1:1 (1-to-1) – Comparing the biometric data from a subject User to the biometric data stored for the expected User.  If the biometric data does not match above the chosen FAR level, the result is a failed match.

1:N (1-to-N) – Comparing the biometric data from one individual to the biometric data from a list of known individuals, the faces of the people on the list that look similar are returned.  This is used for facial recognition surveillance, but can also be used to flag duplicate enrollments.

Artifact (Artefact) –  An inanimate object that seeks to reproduce human biometric traits. 

Authentication – The concurrent Liveness Detection, 3D depth detection, and biometric data verification (i.e., face sharing) of the User.

Bad Actor – A criminal; a person with intentions to commit fraud by deceiving others.

Biometric – The measurement and comparison of data representing the unique physical traits of an individual for the purposes of identifying that individual based on those unique traits.

Certification – The testing of a system to verify its ability to meet or exceed a specified performance standard.  Testing labs Like iBeta issue certifications.

Complicit User Fraud – When a User pretends to have fraud perpetrated against them, but has been involved in a scheme to defraud by stealing an asset and trying to get it replaced by an institution.

Cooperative User – When a testing organization is guided by ISO 30107-3 ISO, the human Subjects used in the tests must provide any and all biometric data that is requested.  This helps to assess the complicit User fraud and phishing risk, but only applies if the test includes matching (not recommended).

Centralized Biometric – Biometric data is collected on any supported device, encrypted and sent to a server for enrollment and later authentication for that device or any other supported device.  When the User’s original biometric data is stored on a secure 3rd-party server, that data can continue to be used as the source of trust and their identity can be established and verified at any time.  Any supported device can be used to collect and send biometric data to the server for comparison, enabling Users to access their accounts from all of their devices, new devices, etc., just like with passwords.  Liveness is the most critical component of a centralized biometric system, and because certified Liveness did not exist until recently, centralized biometrics have not yet been widely deployed.

Credential Sharing – When two or more individuals do not keep their credentials secret and can access each others accounts.  This can be done to subvert licensing fees or to trick an employer into paying for time not worked (also called “buddy punching”).

Credential Stuffing – A cyberattack where stolen account credentials, usually comprising lists of usernames and/or email addresses and the corresponding passwords, are used to gain unauthorized user account access.

Decentralized Biometric – When biometric data is captured and stored on a single device and the data never leaves that device.  Fingerprint readers in smartphones and Apple’s Face ID are examples of decentralized biometrics. They only unlock one specific device, they require re-enrollment on any new device, and further do not prove the identity of the User whatsoever.  Decentralized biometric systems can be defeated easily if a bad actor knows the device's override PIN number, allowing them to overwrite the User’s biometric data with their own.

End User– An individual human who is using an application.

Enrollment – When biometric data is collected for the first time, encrypted and sent to the server.  Note: Liveness must be verified and a 1:N check should be performed against all the other enrollments to check for duplicates.

Face Authentication – Authentication has three parts: Liveness Detection, 3D Depth Detection and Identity Verification.  All must be done concurrently on the same face frames.

Face Matching – Newly captured images/biometric data of a person are compared to the enrolled (previously saved) biometric data of the expected User, determining if they are the same.

Face Recognition – Images/biometric data of a person are compared against a large list of known individuals to determine if they are the same person.

Face Verification – Matching the biometric data of the Subject User to the biometric data of the Expected User.

FAR (False Acceptance Rate) – The probability that the system will accept an imposter’s biometric data as the correct User’s data and incorrectly provide access to the imposter.

FIDO – Stands for Fast IDentity Online:  A standards organization that provides guidance to organization that choose to use Decentralized Biometric Systems (https://fidoalliance.org).

FRR/FNMR/FMR – The probability that a system will reject the correct User when that User’s biometric data is presented to the sensor.  If the FRR is high, Users will be frustrated with the system because they are prevented from accessing their own accounts.

Hill-Climbing Attack – When an attacker uses information returned by the biometric authenticator (match level or liveness score) to learn how to curate their attacks and gain a higher probability of spoofing the system. 

iBeta – A NIST-certified testing lab in Denver Colorado; the only lab currently certifying biometric systems for anti-spoofing/Liveness Detection to the ISO 30107-3 standard (ibeta.com).

Identity & Access Management (IAM) – A framework of policies and technologies to ensure only authorized users have the appropriate access to restricted technology resources, services, physical locations and accounts. Also called identity management (IdM).

Imposter – A living person with traits so similar to the Subject User that the system determines the biometric data is from the same person.

ISO 30107-3 – The International Organization for Standardization’s testing guidance for evaluation of Anti-Spoofing technology (www.iso.org/standard/67381.html).

Knowledge-Based Authentication (KBA) - Authentication method that seeks to prove the identity of someone accessing a digital service. KBA requires knowing a user's private information to prove that the person requesting access is the owner of the digital identity. Static KBA is based on a pre-agreed set of shared secrets. Dynamic KBA is based on questions generated from additional personal information.

Liveness Detection – The ability for a biometric system to determine if data has been collected from a live human or an inanimate, non-living Artifact.

NIST – National Institute of Standards and Technology – The U.S. government agency that provides measurement science, standards, and technology to advance economic advantage in business and government (nist.gov).

Phishing – When a User is tricked into giving a Bad Actor their passwords, PII, credentials, or biometric data.  Example: A User gets a phone call from a fake customer service agent and they request the User’s password to a specific website.

PII – Personally Identifiable Information is information that can be used on its own or with other information to identify, contact, or locate a single person, or to identify an individual in context (en.wikipedia.org/wiki/Personally_identifiable_information).

Presentation Attack Detection (PAD) – A framework for detecting presentation attack events. Related to Liveness Detection and Anti-Spoofing.

Root Identity Provider – An organization that stores biometric data appended to the corresponding personal information of individuals, and allows other organizations to verify the identities of Subject Users by providing biometric data to the Root Identity Provider for comparison.

Spoof – When a non-living object that exhibits some biometric traits is presented to a camera or biometric sensor.  Photos, masks or dolls are examples of Artifacts used in spoofs.

Subject User – The individual that is presenting their biometric data to the biometric sensor at that moment.

Synthetic Identity - When a bad actor uses a combination of biometric data, name, social security number, address, etc. to create a new record for a person who doesn't actually exist, for the purposes of using an account in that name.



Editors & Contributors
 

Kevin Alan Tussy
Editor-in-Chief

LinkedIn

John Wojewidka
Senior Editor

LinkedIn

Josh Rose
Tech Editor

LinkedIn

 
© 2020, Liveness.com. All rights reserved.